The invention relates to gene therapy approaches to supply nucleotide sequences encoding modified forms of the native forms of apolipoprotein A-I (ApoA-I) i.e., mature ApoA-I, preproApoA-I and proApoA-I; ApoA-I peptides; ApoA-I agonists and superagonists, peptides which mimic or exceed the activity of native ApoA-I; and the native ApoA-I gene for the treatment of disorders associated with dyslipoproteinemia, including cardiovascular disease, atherosclerosis, restenosis, hyperlipidemia, and other disorders such as septic shock.
Circulating cholesterol is carried by plasma lipoproteinsxe2x80x94particles of complex lipid and protein composition that transport lipids in the blood. Low density lipoproteins (LDL), and high density lipoproteins (HDL) are the major cholesterol carriers. LDL are believed to be responsible for the delivery of cholesterol from the liver (where it is synthesized or obtained from dietary sources) to extrahepatic tissues in the body. The term xe2x80x9creverse cholesterol transportxe2x80x9d describes the transport of cholesterol from extrahepatic tissues to the liver where it is catabolized and eliminated. It is believed that plasma HDL particles play a major role in the reverse transport process, acting as scavengers of tissue cholesterol.
The evidence linking elevated serum cholesterol to coronary heart disease is overwhelming. For example, atherosclerosis is a slowly progressive disease characterized by the accumulation of cholesterol within the arterial wall. Compelling evidence supports the concept that lipids deposited in atherosclerotic lesions are derived primarily from plasma LDL; thus, LDLs have popularly become known as the xe2x80x9cbadxe2x80x9d cholesterol. In contrast, HDL serum levels correlate inversely with coronary heart diseasexe2x80x94indeed, high serum levels of HDL are regarded as a negative risk factor. It is hypothesized that high levels of plasma HDL are not only protective against coronary artery disease, but may actually induce regression of atherosclerotic plaques (e.g., see Badimon et al., 1992, Circulation 86 (Suppl. III):86-94). Thus, HDL have popularly become known as the xe2x80x9cgoodxe2x80x9d cholesterol.
2.1. Cholesterol Transport
The fat-transport system can be divided into two pathways: an exogenous one for cholesterol and triglycerides absorbed from the intestine, and an endogenous one for cholesterol and triglycerides entering the bloodstream from the liver and other non-hepatic tissue.
In the exogenous pathway, dietary fats are packaged into lipoprotein particles called chylomicrons which enter the bloodstream and deliver their triglycerides to adipose tissue (for storage) and to muscle (for oxidation to supply energy). The remnant of the chylomicron, containing cholesteryl esters, is removed from the circulation by a specific receptor found only on liver cells. This cholesterol then becomes available again for cellular metabolism or for recycling to extrahepatic tissues as plasma lipoproteins.
In the endogenous pathway, the liver secretes a large, very-low-density lipoprotein particle (VLDL) into the bloodstream. The core of VLDLs consists mostly of triglycerides synthesized in the liver, with a smaller amount of cholesteryl esters (either synthesized in the liver or recycled from chylomicrons). Two predominant proteins are displayed on the surface of VLDLs, apoprotein B-100 and apoprotein E. When a VLDL reaches the capillaries of adipose tissue or of muscle, its triglycerides are extracted resulting in a new kind of particle, decreased in size and enriched in cholesteryl esters but retaining its two apoproteins, called intermediate-density lipoprotein (IDL).
In human beings, about half of the IDL particles are removed from the circulation quickly (within two to six hours of their formation), because they bind tightly to liver cells which extract their cholesterol to make new VLDL and bile acids. The IDL particles which are not taken up by the liver remain in the circulation longer. In time, the apoprotein E dissociates from the circulating particles, converting them to LDL having apoprotein B-100 as their sole protein.
Primarily, the liver takes up and degrades most of the cholesterol to bile acids, which are the end products of cholesterol metabolism. The uptake of cholesterol containing particles is mediated by LDL receptors, which are present in high concentrations on hepatocytes. The LDL receptor binds both apoprotein E and apoprotein B-100, and is responsible for binding and removing both IDLs and LDLs from the circulation. However, the affinity of apoprotein E for the LDL receptor is greater than that of apoprotein B-100. As a result, the LDL particles have a much longer circulating life span than IDL particlesxe2x80x94LDLs circulate for an average of two and a half days before binding to the LDL receptors in the liver and other tissues. High serum levels of LDL (the xe2x80x9cbadxe2x80x9d cholesterol) are positively associated with coronary heart disease. For example, in atherosclerosis, cholesterol derived from circulating LDLs accumulates in the walls of arteries leading to the formation of bulky plaques that inhibit the flow of blood until a clot eventually forms, obstructing the artery causing a heart attack or stroke.
Ultimately, the amount of intracellular cholesterol liberated from the LDLs controls cellular cholesterol metabolism. The accumulation of cellular cholesterol derived from VLDLs and LDLs controls three processes: first, it reduces the cell""s ability to make its own cholesterol by turning off the synthesis of HMGCOA reductasexe2x80x94a key enzyme in the cholesterol biosynthetic pathway. Second, the incoming LDL-derived cholesterol promotes storage of cholesterol by activating ACATxe2x80x94the cellular enzyme which converts cholesterol into cholesteryl esters that are deposited in storage droplets. Third, the accumulation of cholesterol within the cell drives a feedback mechanism that inhibits cellular synthesis of new LDL receptors. Cells, therefore, adjust their complement of LDL receptors so that enough cholesterol is brought in to meet their metabolic needs, without overloading. (For a review, see Brown and Goldstein, In, The Pharmacological Basis Of Therapeutics, 8th Ed., Goodman and Gilman, Pergamon Press, NY, 1990, Ch. 36, pp. 874-896).
2.2. Reverse Cholesterol Transport
In sum, peripheral (non-hepatic) cells obtain their cholesterol from a combination of local synthesis and the uptake of preformed sterol from VLDLs and LDLs. In contrast, reverse cholesterol transport (RCT) is the pathway by which peripheral cell cholesterol can be returned to the liver for recycling to extrahepatic tissues, or excretion into the intestine in bile. The RCT pathway represents the only means of eliminating cholesterol from most extrahepatic tissues, and is crucial to maintenance of the structure and function of most cells in the body.
The RCT consists mainly of three steps: (a) cholesterol efflux, the initial removal of cholesterol from various pools of peripheral cells; (b) cholesterol esterification by the action of lecithin:cholesterol acyltransferase (LCAT), preventing a re-entry of effluxed cholesterol into cells; and (c) uptake/delivery of HDL cholesteryl ester to liver cells. The RCT pathway is mediated by HDLS. HDL is a generic term for lipoprotein particles which are characterized by their high density. The main lipidic constituents of HDL complexes are various phospholipids, cholesterol (ester) and triglycerides. The most prominent apolipoprotein components are A-I and A-II which determine the functional characteristics of HDL; furthermore minor amounts of apolipoproteins C-I, C-II, C-III, D, E, J, etc. have been observed. HDL can exist in a wide variety of different sizes and different mixtures of the above-mentioned constituents dependent on the status of remodeling during the metabolic RCT cascade.
The key enzyme involved in the RCT pathway is LCAT. LCAT is produced mainly in the liver and circulates in plasma associated with the HDL fraction. Cholesteryl ester transfer protein (CETP) and another lipid transfer protein, phospholipid transfer protein (PLTP) contribute to further remodeling the circulating HDL population. CETP can move cholesteryl esters made by LCAT to other lipoproteins, particularly ApoB-containing lipoproteins, such as VLDL. HDL triglycerides can be catabolized by the extracellular hepatic triglyceride lipase, and lipoprotein cholesterol is removed by the liver via several mechanisms.
Each HDL particle contains at least one copy (and usually two to four copies) of ApoA-I. ApoA-I is synthesized by the liver and small intestine as preproapolipoprotein which is secreted as a proprotein that is rapidly cleaved to generate a mature polypeptide having 243 amino acid residues. ApoA-I consists mainly of 6 to 8 different 22 amino acid repeats spaced by a linker moiety which is often proline, and in some cases consist of a stretch made up of several residues. ApoA-I forms three types of stable structures with lipids: small, lipid-poor complexes referred to as pre-beta-1 HDL; flattened discoidal particles containing only polar lipids (phospholipid and cholesterol) referred to as pre-beta-2 HDL; and spherical particles containing both polar and nonpolar lipids, referred to as spherical or mature HDL (HDL3 and HDL2). Most HDL in the circulating population contain both ApoA-I and ApoA-II (the second major HDL protein) and are referred to herein as the AI/AII-HDL fraction of HDL. However, the fraction of HDL containing only ApoA-I (referred to herein as the AI-HDL fraction) appear to be more effective in RCT. Certain epidemiologic studies support the hypothesis that the AI-HDL fraction is antiartherogenic. (Parra et al., 1992, Arterioscler. Thromb. 12:701-707; Decossin et al., 1997, Eur. J. Clin. Invest. 27:299-307).
Although the mechanism for cholesterol transfer from the cell surface is unknown, it is believed that the lipid-poor complex, pre-beta-1 HDL is the preferred acceptor for cholesterol transferred from peripheral tissue involved in RCT. Cholesterol newly transferred to pre-beta-1 HDL from the cell surface rapidly appears in the discoidal pre-beta-2 HDL. PLTP may increase the rate of disc formation, but data indicating a role for PLTP in RCT is lacking. LCAT reacts preferentially with discoidal and spherical HDL, transferring the 2-acyl group of lecithin or other phospholipids to the free hydroxyl residue of fatty alcohols, particularly cholesterol to generate cholesteryl esters (retained in the HDL) and lysolecithin. The LCAT reaction requires ApoA-I as activator; i.e., ApoA-I is the natural cofactor for LCAT. The conversion of cholesterol to its ester sequestered in the HDL prevents re-entry of cholesterol into the cell, the result being that cholesteryl esters are destined for removal. Cholesteryl esters in the mature HDL particles in the AI-HDL fraction (i.e., containing ApoA-I and no ApoA-II) are removed by the liver and processed into bile more effectively than those derived from HDL containing both ApoA-I and ApoA-II (the AI/AII-HDL fraction). This may be due, in part, to the more effective binding of AI-HDL to the hepatocyte membrane. The existence of an HDL receptor has been hypothesized, and recently a scavenger receptor, SR-BI, was identified as an HDL receptor. (Acton et al., 1996, Science 271:518-520). The SR-BI is expressed most abundantly in steroidogenic tissues (e.g., the adrenals), and in the liver. (Landshulz et al., 1996, J. Clin. Invest. 98:984-995; Rigotti et al., 1996, J. Biol. Chem. 271:33545-33549).
CETP does not appear to play a major role in RCT, and instead is involved in the metabolism of VLDL- and LDL-derived lipids. However, changes in CETP activity or its acceptors, VLDL and LDL, play a role in xe2x80x9cremodelingxe2x80x9d the HDL population. For example, in the absence of CETP, the HDLs become enlarged particles which are not cleared. (For reviews on RCT and HDLs, see Fielding and Fielding, 1995, J. Lipid Res. 36:211-228; Barrans et al., 1996, Biochem. Biophys. Acta. 1300:73-85; Hirano et al., 1997, Arterioscier. Thromb. Vasc. Biol. 17(b):1053-1059.
2.3. Current Treatments for Lowering Serum Cholesterol
A number of treatments are currently available for lowering serum cholesterol and triglycerides (see, e.g., Brown and Goldstein, supra). However, each has its own drawbacks and limitations in terms of efficacy, side-effects and qualifying patient population.
Bile-acid-binding resins are a class of drugs that interrupt the recycling of bile acids from the intestine to the liver; e.g., cholestyramine (Questran Light(copyright), Bristol-Myers Squibb), and colestipol hydrochloride (Colestid(copyright), The Upjohn company). When taken orally, these positively-charged resins bind to the negatively charged bile acids in the intestine. Because the resins cannot be absorbed from the intestine, they are excreted carrying the bile acids with them. The use of such resins, however, at best only lowers serum cholesterol levels by about 20%, and is associated with gastrointestinal side-effects, including constipation and certain vitamin deficiencies. Moreover, since the resins bind other drugs, other oral medications must be taken at least one hour before or four to six hours subsequent to ingestion of the resin; thus, complicating heart patient""s drug regimens.
The statins are cholesterol lowering agents that block cholesterol synthesis by inhibiting AMGCoA reductasexe2x80x94the key enzyme involved in the cholesterol biosynthetic pathway. The statins, e.g., lovastatin (Mevacor(copyright), Merck and Co., Inc.) and pravastatin (Pravachol(copyright), Bristol-Myers Squibb Co.) are sometimes used in combination with bile-acid-binding resins. The statins significantly reduce serum cholesterol and LDL-serum levels, and slow progression of coronary atherosclerosis. However, serum HDL cholesterol levels are only slightly increased. The mechanism of the LDL lowering effect may involve both reduction of VLDL concentration and induction of cellular expression of LDL-receptor, leading to reduced production and/or increased catabolism of LDLs. Side effects, including liver and kidney dysfunction are associated with the use of these drugs (Physicians Desk Reference, Medical Economics Co., Inc., Montvale, N.J. 1997). Recently, the FDA has approved atorvasatatin (an HMGroA reductase inhibitor developed by Parke-Davis) (Warner Lambert) for the treatment of rare but urgent cases of familial hypercholesterolemia (1995, Scrip 20 (19):10).
Niacin, or nicotinic acid, is a water soluble. vitamin B-complex used as a dietary supplement and antihyperlipidemic agent. Niacin diminishes production of VLDL and is effective at lowering LDL. It is used in combination with bile-acid binding resins. Niacin can increase HDL when used at adequate doses, however, its usefulness is limited by serious side effects when used at such doses.
Fibrates are a class of lipid-lowering drugs used to treat various forms of hyperlipidemia, (i.e., elevated serum triglycerides) which may also be associated with hypercholesterolemia. Fibrates appear to reduce the VLDL fraction and modestly increase HDLxe2x80x94however the effects of these drugs on serum cholesterol is variable. In the United States, fibrates have been approved for use as antilipidemic drugs but have not received approval as hypercholerolemia agents. For example, clofibrate (Atromid-S(copyright), Wyeth-Ayerest Laboratories) is an antilipidemic agent which acts (via an unknown mechanism) to lower serum triglycerides by reducing the VLDL fraction. Although serum cholesterol may be reduced in certain patient subpopulations, the biochemical response to the drug is variable, and is not always possible to predict which patients will obtain favorable results. Atromid-S(copyright) has not been shown to be effective for prevention of coronary heart disease. The chemically and pharmacologically related drug, gemfibrozil (Lopid(copyright), Parke-Davis) is a lipid regulating agent which moderately decreases serum triglycerides and VLDL cholesterol, and increases HDL cholesterolxe2x80x94the HDL2 and HDL3 subfractions as well as both ApoA-I and A-II (i.e., the AI/AII-HDL fraction). However, the lipid response is heterogeneous, especially among different patient populations. Moreover, while prevention of coronary heart disease was observed in male patients between 40-55 without history or symptoms of existing coronary heart disease, it is not clear to what extent these findings can be extrapolated to other patient populations (e.g., women, older and younger males). Indeed, no efficacy was observed in patients with established coronary heart disease. Serious side-effects are associated with the use of fibrates including toxicity such as malignancy, (especially gastrointestinal cancer), gallbladder disease and an increased incidence in non-coronary mortality. These drugs are not indicated for the treatment of patients with high LDL or low HDL as their only lipid abnormality (Physicians"" Desk Reference, 1997, Medical Economics Co., Inc., Montvale, N.J.)
Oral estrogen replacement therapy may be considered for moderate hypercholesterolemia in post-menopausal women. However, increases in HDL may be accompanied with an increase in triglycerides. Estrogen treatment is, of course, limited to a specific patent population (postmenopausal women) and is associated with serious side effects including induction of malignant neoplasms, gall bladder disease, thromboembolic disease, hepatic adenoma, elevated blood pressure, glucose intolerance, and hypercalcemia.
Thus, there is a need to develop safer drugs that are efficacious in lowering serum cholesterol, increasing HDL serum levels, preventing coronary heart disease, and/or treating existing disease.
2.4. ApoA-I as a Target
None of the currently available drugs for lowering cholesterol safely elevate HDL levels and stimulate RCTxe2x80x94most appear to operate on the cholesterol transport pathway, modulating dietary intake, recycling, synthesis of cholesterol, and the VLDL population.
While it is desirable to find drugs that stimulate cholesterol efflux and removal, several potential targets in the RCT existxe2x80x94e.g., LCAT, HDL and its various components (ApoA-I, ApoA-II and phospholipids), LTP, and CETPxe2x80x94and it is not known which target would be most effective at achieving desirable lipoprotein profiles and protective effects. Perturbation of any single component in the RCT pathway ultimately affects the composition of circulating lipoprotein populations, and the efficiency of RCT.
Several lines of evidence based on data obtained in vivo implicate the HDL and its major protein component, ApoA-I, in the prevention of atherosclerotic lesions, and potentially, the regression of plaquesxe2x80x94making these attractive targets for therapeutic intervention. First, an inverse correlation exists between serum ApoA-I (HDL) concentration and atherogenesis in man (Gordon and Rifkind, 1989, N. Eng. J. Med. 321:1311-1316; Gordon et al., 1989, Circulation 79:8-15). Indeed, specific subpopulations of HDL have been associated with a reduced risk for atherosclerosis in humans (Miller, 1987, Amer. Heart 113:589-597; Cheung et al., 1991, Lipid Res. 32:383-394).
Second, animal studies support the protective role of ApoA-I (HDL). Treatment of cholesterol fed rabbits with ApoA-I or HDL reduced the development and progression of plaque (fatty streaks) in cholesterol-fed rabbits. (Koizumi et al., 1988, J. Lipid Res. 29:1405-1415; Badimon et al., 1989, Lab. Invest. 60:455-461; Badimon et al., 1990, J. Clin. Invest. 85:1234-1241). However, the efficacy varied depending upon the source of HDL (Beitz et al., 1992, Prostaglandins, Leukotrienes and Essential Fatty Acids 47:149-152; Mezdour et al., 1995, Atherosclerosis 113:237-246).
Third, direct evidence for the role of ApoA-I was obtained from experiments involving transgenic animals. The expression of the human gene for ApoA-I transferred to mice genetically predisposed to diet-induced atherosclerosis protected against the development of aortic lesions (Rubin et al., 1991, Nature 353:265-267). The ApoA-I transgene was also shown to suppress atherosclerosis in ApoE-deficient mice and in Apo(a) transgenic mice (Paszty et al., 1994, J. Clin. Invest. 94:899-903; Plump et al., 1994, Proc. Natl. Acad. Sci. USA 91:9607-9611, Lin et al., 1994, J. Lipid Res. 35:2263-2266). Similar results were observed in transgenic rabbits expressing human ApoA-I (Duverger, 1996, Circulation 94:713-717; Duverger et al., 1996, Arterioscler. Thromb. Vasc. Biol. 16:1424-1429; Emmanuel et al., 1997, Artheriosclerosis 1035:144), and in transgenic rats where elevated levels of human ApoA-I protected against atherosclerosis and inhibited restenosis following balloon angioplasty (Burkey et al., 1992, Circulation, Supplement I, 86:I-472, Abstract No. 1876; Burkey et al., 1995, J. Lipid Res. 36:1463-1473).
The AI-HDL appear to be more efficient at RCT than the AI/AII-HDL fraction. Studies with mice transgenic for human ApoA-I or ApoA-I and ApoA-II (AI/AII) showed that the protein composition of HDL significantly affects its rolexe2x80x94AI-HDL is more anti-atherogenic than AI/AII-HDL (Schultz et al., 1993, Nature 365:762-764). Parallel studies involving transgenic mice expressing the human LCAT gene demonstrate that moderate increases in LCAT activity significantly change lipoprotein cholesterol levels, and that LCAT has a significant preference for HDL containing ApoA-I (Francone et al., 1995, J. Clinic. Invest. 96:1440-1448; Beard et al., 1997, Nature Medicine 3,7:744-749). While these data support a significant role for ApoA-I in activating LCAT and stimulating RCT, additional studies demonstrate a more complicated scenario: a major component of HDL that modulates efflux of cell cholesterol is the phospholipid (Fournier et al., 1996, J. Lipid Res. 37:1704-1711).
In view of the potential role of HDL, i.e., both ApoA-I and its associated phospholipid, in the protection against atherosclerotic disease, human clinical trials utilizing recombinantly produced ApoA-I were commenced, discontinued and apparently recommenced by UCB Belgium (Pharma Projects, Oct. 27, 1995; IMS RandD Focus, Jun. 30, 1997; Drug Status Update, 1997, Atherosclerosis 2(6):261-265); see also M. Eriksson at Congress, xe2x80x9cThe Role of HDL in Disease Prevention,xe2x80x9d Nov. 7-9, 1996, Fort Worth; Lacko and Miller, 1997, J. Lipid Res. 38:1267-1273; and WO94/13819) and were commenced and discontinued by Bio-Tech (Pharmaprojects, Apr. 7, 1989). Trials were also attempted using ApoA-I to treat septic shock (Opal, xe2x80x9cReconstituted HDL as a Treatment Strategy for Sepsis,xe2x80x9d IBCs 7th Annual International Conference on Sepsis, Apr. 28-30, 1997, Washington, D.C.; Gouni et al., 1993, J. Lipid Res. 94:139-146; Levine WO96/04916). However, there are many pitfalls associated with the production and use of ApoA-I making it less than ideal as a drug; e.g., ApoA-I is a large protein that is difficult and expensive to produce; significant manufacturing and reproducibility problems must be overcome with respect to stability during storage, delivery of an active product and half-life in vivo.
In view of these drawbacks, attempts have been made to prepare peptides that mimic ApoA-I. Since the key activities of ApoA-I have been attributed to the presence of multiple repeats of a unique secondary structural feature in the proteinxe2x80x94a class A amphipathic xcex1-helix (Segrest, 1974, FEBS Lett. 38:247-253), most efforts to design peptides which mimic the activity of ApoA-I have focused on designing peptides which form class A-type amphipathic xcex1-helices.
Class A-type amphipathic xcex1-helices are unique in that positively charged amino acid residues are clustered at the hydrophobic-hydrophilic interface and negatively charged amino acid residues are clustered at the center of the hydrophilic face. Furthermore, Class A xcex2-helical peptides have a hydrophobic angle of less than 180xc2x0 (Segrest et al., PROTEINS, 1990; Structure, Function and Genetics 8:103-117). The initial de novo strategies to design ApoA-I mimics were not based upon the primary sequences of naturally occurring apolipoproteins, but rather upon incorporating these unique Class A helix features into the sequences of the peptide analogues as well as some of the properties of the ApoA-I domains (see, e.g., Davidson et al., 1996, Proc. Natl. Acad. Sci. USA 93:13605-13610; Rogers et al., 1997, Biochemistry 36:288-300; Lins et al., 1993, Biochim. Biophys. Acta biomembranes 1151:137-142; Ji and Jonas, 1995, J. Biol. Chem. 270:11290-11297; Collet et al., 1997, Journal of Lipid Research, 38:634-644; Sparrow and Gotto, 1980, Ann. N.Y. Acad. Sci. 348:187-211; Sparrow and Gotto, 1982, CRC Crit. Rev. Biochem. 13:87-107; Sorci-Thomas et al., 1993, J. Biol. Chem. 268:21403-21409; Wang et al., 1996, Biochim. Biophys.
Acta 174-184; Minnich et al., 1992, J. Biol. Chem. 267:16553-16560; Holvoet et al., 1995, Biochemistry 34:13334-13342; Sorci-Thomas et al., 1997, J. Biol. Chem. 272 (11):7278-7284; and Frank et al., 1997, Biochemistry 36:1798-1806).
In one study, Fukushima et al. synthesized a 22-residue peptide composed entirely of Glu, Lys and Leu residues arranged periodically so as to form an amphipathic xcex1-helix with equal hydrophilic and hydrophobic faces (xe2x80x9cELK peptidexe2x80x9d) (Fukushima et al., 1980, J. Biol. Chem. 255:10651-10657; Fukushima et al., 1979, J. Amer. Chem. Soc. 101(13):3703-3704). The ELK peptide shares 41% sequence homology with the 198-219 fragment of ApoA-I. As studied by quantitative ultrafiltration, gel permeation chromatography and circular dichroism, this ELK peptide was shown to effectively associate with phospholipids and mimic some of the physical and chemical properties of ApoA-I (Kaiser et al., 1983, Proc. Natl. Acad. Sci. USA 80:1137-1140; Kaiser et al., 1984, Science 223:249-255; Fukushima et al., 1980, supra; Nakagawa et al., 1985, J. Am. Chem. Soc. 107:7087-7092). Yokoyama et al. concluded from such studies that the crucial factor for LCAT activation is simply the presence of a large enough amphipathic structure (Yokoyama et al., 1980, J. Biol. Chem. 255(15):7333-7339). A dimer of this 22-residue peptide was later found to more closely mimic ApoA-I than the monomer; based on these results, it was suggested that the 44-mer which is punctuated in the middle by a helix breaker (either Gly or Pro), represented the minimal functional domain in ApoA-I (Nakagawa et al., 1985, supra).
Another study involved model amphipathic peptides called xe2x80x9cLAP peptidesxe2x80x9d (Pownall et al., 1980, Proc. Natl. Acad. Sci. USA 77(b):3154-3158; Sparrow et al., 1981, In: Peptides:Synthesis-Structure-Function, Roch and Gross, Eds., Pierce Chem. Co., Rockford, Ill., 253-256). Based on lipid binding studies with fragments of native apolipoproteins, several LAP peptides were designed, designated LAP-16, LAP-20 and LAP-24 (containing 16, 20 and 24 amino acid residues, respectively). These model amphipathic peptides share no sequence homology with the apolipoproteins and were designed to have polar faces organized in a manner unlike the class A-type amphipathic helical domains associated with apolipoproteins. From these studies, the authors concluded that a minimal length of 20 residues is necessary to confer lipid-binding properties to model amphipathic peptides.
Studies with mutants of LAP20 containing a proline residue at different positions in the sequence indicated that a direct relationship exists between lipid binding and LCAT activation, but that the helical potential of a peptide alone does not lead to LCAT activation (Ponsin et al., 1986 J. Biol. Chem. 261(20):9202-9205). Moreover, the presence of this helix breaker (Pro) close to the middle of the peptide reduced its affinity for phospholipid surfaces as well as its ability to activate LCAT. While certain of the LAP peptides were shown to bind phospholipids (sparrow et al., supra), controversy exists as to the extent to which LAP peptides are helical in the presence of lipids (Buchko et al., 1996, J. Biol. Chem. 271(6):3039-3045; Zhong et al., 1994, Peptide Research 7(2):99-106).
Segrest et al. have synthesized peptides composed of 18 to 24 amino acid residues that share no sequence homology with the helices of ApoA-I (Kannelis et al., 1980, J. Biol. Chem. 255(3):11464-11472; Segrest et al., 1983, J. Biol. Chem. 258:2290-2295). The sequences were specifically designed to mimic the amphipathic helical domains of class A exchangeable apolipoproteins in terms of hydrophobic moment (Eisenberg et al., 1982, Nature 299:371-374) and charge distribution (Segrest et al., 199.0, Proteins 8:103-117; U.S. Pat. No. 4,643,988). One 18-residue peptide, the xe2x80x9c18Axe2x80x9d peptide, was designed to be a model class-A xcex1-helix (Segrest et al., 1990, supra). Studies with these peptides and other peptides having a reversed charged distribution, like the xe2x80x9c18Rxe2x80x9d peptide, have consistently shown that charge distribution is critical for activity; peptides with a reversed charge distribution exhibit decreased lipid affinity relative to the 18A class-A mimics and a lower helical content in the presence of lipids (Kanellis et al., 1980, J. Biol. Chem. 255:11464-11472; Anantharamaiah et al., 1985, J. Biol. Chem. 260:10248-10255; Chung et al., 1985, J. Biol. Chem. 260:10256-10262; Epand et al., 1987, J. Biol. Chem. 262:9389-9396; Anantharamaiah et al., 1991, Adv. Exp. Med. Biol. 285:131-140).
Other synthetic peptides sharing no sequence homology with the apolipoproteins which have been proposed with limited success include dimers and trimers of the 18A peptide (Anatharamaiah et al., 1986, Proteins of Biological Fluids 34:63-66), GALA and EALA peptides (Subbarao et al., 1988, Proteins: Structure, Function and Genetics 3:187-198) and ID peptides (Labeur et al., 1997, Arteriosclerosis, Thrombosis, and Vascular Biology 17:580-588) and the 18AM4 peptide (Brasseur et al., 1993, Biochem. Biophys. Acta 1:170:1-7).
A xe2x80x9cconsensusxe2x80x9d peptide containing 22-amino acid residues based on the sequences of human ApoA-I has also been designed (Venkatachalapathi et al., 1991, Mol. Conformation and Bio. Interactions B:585-596). The sequence was constructed by identification of the most prevalent residue at each position of the helices of ApoA-I. Like the peptides described above, the helix formed by this peptide has positively charged amino acid residues clustered at the hydrophilic:hydrophobic interface, negatively charged amino acid residues clustered at the center of the hydrophilic face and hydrophobic angle of less than 180xc2x0. While a dimer of this peptide is somewhat effective in activating LCAT, the monomer exhibited poor lipid binding properties (Vekatachalapathi et al., 1991, supra).
Based primarily on in vitro studies with the peptides described above, a set of xe2x80x9crulesxe2x80x9d has emerged for designing peptides which mimic the function of ApoA-I. Significantly, it is thought that an amphipathic xcex1-helix having positively charged residues clustered at the hydrophilic-hydrophobic interface and negatively charged amino acid residues clustered at the center of the polar face is required for lipid affinity and LCAT activation (Venkatachalapathi et al., 1991, supra). Anantharamaiah et al., have also indicated that the negatively charged Glu residue at position 13 of the consensus 22-mer peptide, which is positioned within the hydrophobic face of the xcex1-helix, plays an important role in LCAT activation (Anantharamaiah et al., 1991, supra). Furthermore, Brasseur has indicated that a hydrophobic angle (pho angle) of less than 180xc2x0 is required for optimal lipid-apolipoprotein complex stability, and also accounts for the formation of discoidal particles having the peptides around the edge of a lipid bilayer (Brasseur, 1991, J. Biol. Chem. 66(24):16120-16127). Rosseneu et al., have also insisted that a hydrophobic angle of less than 180xc2x0 is required for LCAT activation (WO93/25581).
However, despite these xe2x80x9crulesxe2x80x9d to date, no one has designed or produced a peptide as active as ApoA-Ixe2x80x94the best having less than 40% of the activity of ApoA-I as measured by the LCAT activation assay described herein. None of the peptide xe2x80x9cmimeticsxe2x80x9d described in the literature have not been demonstrated to be useful as a drug.
In view of the foregoing, there is a need for the development of a stable ApoA-I agonist that mimics the activity of ApoA-I, which is relatively simple and cost-effective to produce. However, the xe2x80x9crulesxe2x80x9d for designing efficacious ApoA-I mimetics have not been unraveled and the principles to devise organic molecules with the function of ApoA-I are unknown.
The invention relates to genetic approaches to supply nucleotide sequences encoding modified forms of the native forms of apolipoprotein A-I (ApoA-I): mature ApoA-I, preproApoA-I and proApoA-I; including native ApoA-I modified to contain ApoA-I agonists, peptides which mimic the activity of ApoA-I; ApoA-I superagonists, peptides which exceed the activity of native ApoA-I; and modified native ApoA-I having one or more amphipathic helices replaced by the nucleotide sequences of one or more ApoA-I agonists; for the treatment of disorders associated with dyslipoproteinemia, including cardiovascular disease, atherosclerosis, restenosis, hyperlipidemia, and other disorders such as septic shock.
In particular, the invention relates to genetic approaches to supply nucleotide sequences encoding modified native forms of the ApoA-I protein and peptides which act as ApoA-I agonists capable of forming amphipathic xcex1-helices (in the presence of lipids) that approach or surpass the activity of native ApoA-I, e.g., formation of pre-xcex2-like or HDL-like complexes, promotion of cholesterol efflux, binding to lipids, increasing the activity of LCAT, and increasing serum levels of HDL.
The present invention also relates to nucleotide sequences encoding a modified native ApoA-I having one or more amphipathic helices replaced by the nucleotide sequences of one or more ApoA-I agonists.
The present invention relates to DNA vectors or cassettes that contain nucleotide sequences encoding modified ApoA-I or peptides which act as ApoA-I agonists or superagonists. The invention further relates to DNA vectors or cassettes that contain nucleotide sequences encoding modified forms of the native ApoA-I gene. The DNA vectors or cassettes may encode modified forms of native ApoA-I or ApoA-I agonists under the control of strong regulatory elements which result in high levels of ApoA-I expression. To increase efficiency of production, the DNA expression vectors or cassettes may be designed to encode multiple units of the peptide, e.g., the cassettes may contain nucleotide sequences encoding several ApoA-I peptides in tandem, separated by an internal ribosome binding site, so that several peptides may be encoded by the same cassette. The vectors or cassettes may also express ApoA-I peptides as monopolymers (repeating peptide units) or heteropolymers (different peptide units). The cassettes of the present invention may also additionally encode prepropeptide or propeptide ApoA-I sequences so that the peptides of the present invention are processed in the same way as the native ApoA-I protein.
The present invention further relates to genetically engineered host cells which express the modified native forms of ApoA-I: mature ApoA-I, preproform or proform, peptides which mimic ApoA-Iactivity or the native ApoA-I gene product. The host cells may be genetically engineered in vitro or in vivo. The present invention also relates to transgenic animals engineered to express the ApoA-I peptides of the present invention.
The invention also relates to in vivo delivery applications for gene therapy, including administering engineered viral vectors or liposomes which contain DNA sequences encoding ApoA-I or ApoA-I agonist peptides in monomeric or heteromeric forms. The invention relates to ex vivo gene therapy approaches to provide engineered cells derived from a patient, or a compatible host, to express modified forms of the native ApoA-I gene, ApoA-I agonists in monomeric or heteromeric form. Ex vivo gene therapy approaches may also encompass delivery of xe2x80x9csacs of cellsxe2x80x9d which have been genetically engineered to express the ApoA-I gene of the present invention, including, ApoA-I agonists in monomeric or heteromeric form and additional ApoA-I prosequences. In a preferred embodiment, human hepatocytes may be engineered to express ApoA-I agonists.
The invention is based, in part, on the Applicants"" design and discovery of peptides that mimic the helical structure and amphipathic properties of the helical amphipathic domains of ApoA-I. Surprisingly, the peptides of the invention have a ApoA-I activity, i.e., formation of HDL-like complexes and promotion of cholesterol efflux, well above those reported for ApoA-I-derived peptides described in the literature. Indeed, some embodiments of the invention approach 100% of the activity of native ApoA-I, whereas superagonists described herein exceed the specific activity of ApoA-I.
The invention is illustrated by way of working examples that describe the in vitro and in vivo activity and efficacy of the ApoA-I peptides and agonists encoded by the nucleic acids of the present invention. Based upon the structure and activity of the exemplified embodiments, the Applicants have devised a set of xe2x80x9crulesxe2x80x9d which can be used to design altered or mutated forms that are also within the scope of the invention.
The invention also relates to pharmaceutical formulations containing such cell lines and vectors encoding modified forms of the native ApoA-I, i.e., mature ApoA-I, preproform of ApoA-I or proform of ApoA-I, or ApoA-I agonists (either as peptides or peptide-lipid complexes) as the active ingredient, as well as methods for preparing such formulations and their use to treat diseases associated with dyslipoproteinemia (e.g., cardiovascular diseases, atherosclerosis), restenosis, hyperlipidemia, hypertriglyceridemia, and/or endotoxinemia (e.g., septic shock).
3.1. Abbreviations
As used herein, the abbreviations for the genetically encoded amino acids are as follows:
3.2. Definitions
As used herein, the following terms shall have the following meanings:
xe2x80x9cHydrophophilic Amino Acidxe2x80x9d refers to an amino acid having a side chain that tends to associate with the aqueous phase or polar head of a phospholipid when in the presence of lipids. Genetically encoded hydrophilic amino acids include Asp (D), Glu (E), His (H), Lys (K), Arg (R), Asn (N), Gln (Q), Ser (S) and Thr (T).
xe2x80x9cAcidic Amino Acidxe2x80x9d refers to an amino acid having a side chain pK value of less than 7. Acidic amino acids typically have negatively charged side chains at physiological pH due to loss of a hydrogen ion. Genetically encoded acidic amino acids include Glu (E) and Asp (D).
xe2x80x9cBasic Amino Acidxe2x80x9d refers to an amino acid having a side chain pK value of greater than 7. Basic amino acids typically have positively charged side chains at physiological pH due to association with hydronium ion. Genetically encoded basic amino acids include Arg (R) and Lys (K).
xe2x80x9cPolar Amino Acidxe2x80x9d refers to an amino acid having a side chain that is uncharged at physiological pH, but which has at least one bond wherein the pair of electrons shared in common by two atoms is held more closely by one of the atoms. Genetically encoded polar amino acids include Asn (N) and Gln (Q).
xe2x80x9cHydrophobic Amino Acidxe2x80x9d refers to an amino acid having a side chain that tends to associate with the acyl chains of a phospholipid when in the presence of lipids. Genetically encoded hydrophobic amino acids include Phe (F), Tyr (Y), Trp (W), Leu (L), Val (V), Ile (I), Met (M), Gly (G), Ala (A) and Pro (P).
xe2x80x9cAromatic Amino Acidxe2x80x9d refers to a hydrophobic amino acid with a side chain having at least one aromatic or heteroaromatic ring. The aromatic or heteroaromatic ring may contain one or more substituents. Genetically encoded aromatic amino acids include Phe (F), Tyr (Y) and Trp (W).
xe2x80x9cNonpolar Amino Acidxe2x80x9d refers to a hydrophobic amino acid having a side chain that is uncharged at physiological pH and which has bonds wherein the pair of electrons shared in common by two atoms is generally held equally by each at the two atoms (i.e., the side chain is not polar).
xe2x80x9cAliphatic Amino Acidxe2x80x9d refers to a hydrophobic amino acid having an aliphatic hydrocarbon side chain. Genetically encoded aliphatic amino acids include Ala (A), Val (V), Leu (L) and Ile (I).